Species differences, but not habitat, influence catch rate hyperstability across a recreational fishery landscape

Highlights

• Electrofishing catch rates are good proxies of fish abundance.

• Angler catch rates are not proportional to abundance for several sport fish.

• Species differences matter in the degree of vulnerability to catch-rate hyperstability.

• Habitat did not appear to influence catch-rate hyperstability.

• Managers may need to consider variation in the magnitude of hyperstability amongst fisheries.

Abstract

In commercial and recreational fisheries, catch rate is often assumed to be proportional to stock size and is used by managers and fishers as an indicator of fishery sustainability. If catch rate is proportional to stock size, it can signal a decline of stocks and managers can impose restrictive harvest policies or recreational anglers can move to a new system and allow the over-exploited system to rebound. A growing literature has documented catch rates remaining high even as fish stocks decline (i.e., hyperstability of catch rates) leading to delayed management intervention and overexploitation. Although recent evidence has indicated the presence of hyperstability of catch rates in recreational fisheries, whether hyperstability differs across species or system types remains unknown. To investigate whether catch rate hyperstability varies amongst species or systems, we first tested whether electrofishing catch per unit effort (efCPUE) was an appropriate proxy for true abundance. We then compared the relationship between angler catch rate and fish abundance for common freshwater sport fishes across gradients of habitat availability. We found significant differences in the strength of hyperstability amongst species. We did not identify a consistent influence of habitat on hyperstability of catch rates. Angler preferences and behavior may explain some of the variance in non-proportional catch rates. Future research investigating angler behavior, population size structure, and population dynamics in these systems may identify key interactions that create differences in vulnerability to population collapse.

Introduction

Sustainable management of recreational fishery landscapes is challenging as a result of complex interactions between biotic factors (e.g., competition, predation, habitat) and social dynamics, including variation in angler avidity, skill level, and desires (Hickley and Tompkins, 1998, Arlinghaus et al., 2002, Post et al., 2002, Carpenter and Brock, 2004, Lewin et al., 2006, Johnston et al., 2010, Dedual et al., 2013, Post, 2013, Ward et al., 2016, Embke et al., 2019, Solomon et al., 2020). One major challenge for managers of recreational fisheries is the relationship between fish abundance and angler catch rates. In the simplest scenario, angler catch rates and fish abundance are proportional leading to self-regulating fisheries when catch rates drive angler decision making. If catch rates are proportional to fish abundance, declines in catch rates would signal declines in abundance to anglers who may subsequently move to other waterbodies (if available) where catch rates are expected to be higher; with reduced effort, fish abundance is expected to rebound through natural recruitment and reduced fishing mortality. However, if catch rates are hyperstable and do not decline proportionally with abundance, anglers may continue to catch fish at a high rate despite low abundances and over-exploit, or even collapse, the fishery (Harley et al., 2001, Post et al., 2002, Ward et al., 2013).

Although hyperstable catch rates have been known to exist in temperate marine commercial fisheries for decades (Creco and Overholtz, 1990; Hilborn and Walters, 1992; Harley et al., 2001), documentation of these patterns and the mechanisms underpinning hyperstability of catch rates in recreational fisheries has only emerged over the last twenty years (Hansen et al., 2005, Erisman et al., 2011; Ward et. al, 2013; Maggs et al., 2016; Mrnak et al., 2018; Dassow et al., 2020; Feiner et al., 2020). To this point, three distinct mechanisms have been invoked as drivers of hyperstability in recreational fisheries. First, targeting of aggregated fish is a well-known mechanism that can cause hyperstable catch rates. For example, aggregation during spawning was thought to drive hyperstable catch rates in two species of marine bass (Paralabrax spp., Erisman et al., 2011) and angler targeting of preferred habitat was thought to explain hyperstable catch rates in largemouth bass (Micropterus salmoides, Dassow et al., 2020). Second, low-skill anglers can leave a fishery as their success declines with fish abundance leaving high-skill anglers that maintain high catch rates even at low fish abundance. This second mechanism is referred to as “effort sorting” and has been identified as a driver of hyperstable catch rates in recreational fisheries targeting rainbow trout (Oncorhynchus mykiss) in British Columbia (Ward et al., 2013, van Poorten et al., 2016). Finally, advancements in angler technology have also been shown to produce hyperstable catch rates in recreational fisheries over longer time scales, as was observed in the competitive shoreline fishery for leerfish (Lichia amia) in South Africa (Maggs et al., 2016).

Because catch, and consequently hyperstability, arises from interactions between fish, habitat, and fishers, it might be expected that species identity and habitat characteristics influence the strength of hyperstability. Differences in average abundance, body size, habitat preferences, and foraging behavior amongst recreationally targeted fish species could likely generate inter-specific variance in recreational angler catch rates. In addition, differences in behavior, skill, or investment in technology of anglers targeting different fishes may alter the degree to which catch rates are hyperstable amongst species. Contrary to this expectation, Harley et al. (2001) found that the strength of hyperstability was largely similar among ten species in multiple North Atlantic marine commercial fishery regions. Beyond the research of Harley et al. (2001), few studies have attempted to quantify hyperstability in more than one fishery at a time (Erisman et al., 2011, Ward et al., 2013, Maggs et al., 2016, Mrnak et al., 2018, Dassow et al., 2020, but see Feiner et al., 2020). As a result, we hypothesized that, consistent with the findings of Harley et al. (2001), we would not observe differences in the strength of hyperstability amongst species in inland lake recreational fisheries.

Given the prominence of aggregation as a mechanism underpinning hyperstability of catch rates and the influence of habitat on fish behavior, we expect habitat availability to be a likely driver of variation in hyperstability among lakes. However, research investigating the roles of habitat preference and aggregation as mediators of hyperstability has yet to quantify the strength or spatial scale of aggregation required to generate hyperstable catch rates (Erisman et al., 2011, Dassow et al., 2020). Therefore, variation in habitat at multiple spatial scales could modify the strength of hyperstability. For example, lake surface area may influence the strength of hyperstability because small lakes allow anglers to more easily target and capture individuals. Conversely in large lakes, aggregations of fish may be harder to locate at low fish abundances leading to lower catch rates and thus a less hyperstable, and more proportional, relationship between catch rates and fish abundance. In addition, for certain species the availability of shoreline habitat might drive proportional or non-linear changes in the strength of hyperstability. Finally, availability of coarse woody habitat for refuge and ambush sites may alter fish aggregation and angler behavior at fine spatial scales and modify the strength of hyperstability across lakes (Sass et al., 2006a; Pine et al., 2009; Ziegler et al., 2019). Given the importance of habitat for fish behavior, growth, and reproduction (Schindler et al., 2000, Sass et al., 2006b, Gaeta et al., 2011, Gaeta et al., 2014), we hypothesized that habitat, at one or more spatial scales, would be a strong modifier of the strength of hyperstability across lakes in a recreational fishery landscape. Specifically, we predicted that at low habitat availability fish would aggregate most strongly and produce the most hyperstable angler catch rates.

To test whether differences in species’ ecology or lake-to-lake variation in habitat availability influences the magnitude of recreational angler catch rate hyperstability, we used a comparative approach that leveraged long-term, regional fish relative abundance and recreational angler catch rate data for six sport fishes in Wisconsin lakes. Previous research has used the exponent of a power-law relationship between catch rates and fish abundance (β) to quantify the strength of hyperstability of catch rates (Ward et al., 2013). We extended this approach by testing whether β varies as a function of species or indicators of lake habitat availability (e.g., lake surface area, shoreline complexity) using a multiple model comparison framework. If models that include species identity or information about habitat outperform the traditional power-law model, we would infer that these factors modify the magnitude of catch rate hyperstability amongst fisheries. A better understanding of what factors influence catch rate hyperstability could then enhance our ability to model fishery dynamics and our ability to manage these systems.